Machine Learning Approach for Biopsy-Based Identification of Eosinophilic Esophagitis Reveals Importance of Global features

<italic>Goal:</italic> Eosinophilic esophagitis (EoE) is an allergic inflammatory condition characterized by eosinophil accumulation in the esophageal mucosa. EoE diagnosis includes a manual assessment of eosinophil levels in mucosal biopsies&#x2013;a time-consuming, laborious task t...

Full description

Saved in:
Bibliographic Details
Main Authors: Tomer Czyzewski, Nati Daniel, Mark Rochman, Julie Caldwell, Garrett Osswald, Margaret Collins, Marc Rothenberg, Yonatan Savir
Format: Article
Language:English
Published: IEEE 2021-01-01
Series:IEEE Open Journal of Engineering in Medicine and Biology
Subjects:
Online Access:https://ieeexplore.ieee.org/document/9457060/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849417763969302528
author Tomer Czyzewski
Nati Daniel
Mark Rochman
Julie Caldwell
Garrett Osswald
Margaret Collins
Marc Rothenberg
Yonatan Savir
author_facet Tomer Czyzewski
Nati Daniel
Mark Rochman
Julie Caldwell
Garrett Osswald
Margaret Collins
Marc Rothenberg
Yonatan Savir
author_sort Tomer Czyzewski
collection DOAJ
description <italic>Goal:</italic> Eosinophilic esophagitis (EoE) is an allergic inflammatory condition characterized by eosinophil accumulation in the esophageal mucosa. EoE diagnosis includes a manual assessment of eosinophil levels in mucosal biopsies&#x2013;a time-consuming, laborious task that is difficult to standardize. One of the main challenges in automating this process, like many other biopsy-based diagnostics, is detecting features that are small relative to the size of the biopsy. <italic>Results:</italic> In this work, we utilized hematoxylin- and eosin-stained slides from esophageal biopsies from patients with active EoE and control subjects to develop a platform based on a deep convolutional neural network (DCNN) that can classify esophageal biopsies with an accuracy of 85&#x0025;, sensitivity of 82.5&#x0025;, and specificity of 87&#x0025;. Moreover, by combining several downscaling and cropping strategies, we show that some of the features contributing to the correct classification are global rather than specific, local features. <italic>Conclusions:</italic> We report the ability of artificial intelligence to identify EoE using computer vision analysis of esophageal biopsy slides. Further, the DCNN features associated with EoE are based on not only local eosinophils but also global histologic changes. Our approach can be used for other conditions that rely on biopsy-based histologic diagnostics.
format Article
id doaj-art-0811d8686b0d4e79a1e924565ba7e174
institution Kabale University
issn 2644-1276
language English
publishDate 2021-01-01
publisher IEEE
record_format Article
series IEEE Open Journal of Engineering in Medicine and Biology
spelling doaj-art-0811d8686b0d4e79a1e924565ba7e1742025-08-20T03:32:40ZengIEEEIEEE Open Journal of Engineering in Medicine and Biology2644-12762021-01-01221822310.1109/OJEMB.2021.30895529457060Machine Learning Approach for Biopsy-Based Identification of Eosinophilic Esophagitis Reveals Importance of Global featuresTomer Czyzewski0Nati Daniel1https://orcid.org/0000-0002-0939-3379Mark Rochman2Julie Caldwell3Garrett Osswald4Margaret Collins5Marc Rothenberg6Yonatan Savir7https://orcid.org/0000-0002-5345-8491Department of Physiology, Biophysics, and System Biology, Faculty of Medicine, Technion Israel Institute of Technology, Haifa, IsraelDepartment of Physiology, Biophysics, and System Biology, Faculty of Medicine, Technion Israel Institute of Technology, Haifa, IsraelDivision of Allergy, and Immunology, Cincinnati Children's Hospital Medical Center, Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, USADivision of Allergy, and Immunology, Cincinnati Children's Hospital Medical Center, Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, USADivision of Allergy, and Immunology, Cincinnati Children's Hospital Medical Center, Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, USADivision of Pathology, Cincinnati Children's Hospital Medical Center, Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, USADivision of Allergy, and Immunology, Cincinnati Children's Hospital Medical Center, Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, USADepartment of Physiology, Biophysics, and System Biology, Faculty of Medicine, Technion Israel Institute of Technology, Haifa, Israel<italic>Goal:</italic> Eosinophilic esophagitis (EoE) is an allergic inflammatory condition characterized by eosinophil accumulation in the esophageal mucosa. EoE diagnosis includes a manual assessment of eosinophil levels in mucosal biopsies&#x2013;a time-consuming, laborious task that is difficult to standardize. One of the main challenges in automating this process, like many other biopsy-based diagnostics, is detecting features that are small relative to the size of the biopsy. <italic>Results:</italic> In this work, we utilized hematoxylin- and eosin-stained slides from esophageal biopsies from patients with active EoE and control subjects to develop a platform based on a deep convolutional neural network (DCNN) that can classify esophageal biopsies with an accuracy of 85&#x0025;, sensitivity of 82.5&#x0025;, and specificity of 87&#x0025;. Moreover, by combining several downscaling and cropping strategies, we show that some of the features contributing to the correct classification are global rather than specific, local features. <italic>Conclusions:</italic> We report the ability of artificial intelligence to identify EoE using computer vision analysis of esophageal biopsy slides. Further, the DCNN features associated with EoE are based on not only local eosinophils but also global histologic changes. Our approach can be used for other conditions that rely on biopsy-based histologic diagnostics.https://ieeexplore.ieee.org/document/9457060/Decision support systemdeep convolutional networkdigital pathologyeosinophilic esophagitissmall features detection
spellingShingle Tomer Czyzewski
Nati Daniel
Mark Rochman
Julie Caldwell
Garrett Osswald
Margaret Collins
Marc Rothenberg
Yonatan Savir
Machine Learning Approach for Biopsy-Based Identification of Eosinophilic Esophagitis Reveals Importance of Global features
IEEE Open Journal of Engineering in Medicine and Biology
Decision support system
deep convolutional network
digital pathology
eosinophilic esophagitis
small features detection
title Machine Learning Approach for Biopsy-Based Identification of Eosinophilic Esophagitis Reveals Importance of Global features
title_full Machine Learning Approach for Biopsy-Based Identification of Eosinophilic Esophagitis Reveals Importance of Global features
title_fullStr Machine Learning Approach for Biopsy-Based Identification of Eosinophilic Esophagitis Reveals Importance of Global features
title_full_unstemmed Machine Learning Approach for Biopsy-Based Identification of Eosinophilic Esophagitis Reveals Importance of Global features
title_short Machine Learning Approach for Biopsy-Based Identification of Eosinophilic Esophagitis Reveals Importance of Global features
title_sort machine learning approach for biopsy based identification of eosinophilic esophagitis reveals importance of global features
topic Decision support system
deep convolutional network
digital pathology
eosinophilic esophagitis
small features detection
url https://ieeexplore.ieee.org/document/9457060/
work_keys_str_mv AT tomerczyzewski machinelearningapproachforbiopsybasedidentificationofeosinophilicesophagitisrevealsimportanceofglobalfeatures
AT natidaniel machinelearningapproachforbiopsybasedidentificationofeosinophilicesophagitisrevealsimportanceofglobalfeatures
AT markrochman machinelearningapproachforbiopsybasedidentificationofeosinophilicesophagitisrevealsimportanceofglobalfeatures
AT juliecaldwell machinelearningapproachforbiopsybasedidentificationofeosinophilicesophagitisrevealsimportanceofglobalfeatures
AT garrettosswald machinelearningapproachforbiopsybasedidentificationofeosinophilicesophagitisrevealsimportanceofglobalfeatures
AT margaretcollins machinelearningapproachforbiopsybasedidentificationofeosinophilicesophagitisrevealsimportanceofglobalfeatures
AT marcrothenberg machinelearningapproachforbiopsybasedidentificationofeosinophilicesophagitisrevealsimportanceofglobalfeatures
AT yonatansavir machinelearningapproachforbiopsybasedidentificationofeosinophilicesophagitisrevealsimportanceofglobalfeatures